U.S. patent application number 14/326418 was filed with the patent office on 2015-01-08 for system and method for iterative compensation for linear and nonlinear interference in system employing ftn symbol transmission rates.
The applicant listed for this patent is Hughes Network Systems, LLC. Invention is credited to Bassel BEIDAS, Mustafa EROZ, Lin-Nan LEE, Rohit Iyer SESHADRI.
Application Number | 20150010118 14/326418 |
Document ID | / |
Family ID | 52132828 |
Filed Date | 2015-01-08 |
United States Patent
Application |
20150010118 |
Kind Code |
A1 |
BEIDAS; Bassel ; et
al. |
January 8, 2015 |
SYSTEM AND METHOD FOR ITERATIVE COMPENSATION FOR LINEAR AND
NONLINEAR INTERFERENCE IN SYSTEM EMPLOYING FTN SYMBOL TRANSMISSION
RATES
Abstract
An approach for increasing transmission throughput of a
non-linear wireless channel, and efficient decoding of the
transmitted signal via a simplified receiver, is provided. A signal
reflects a source signal, and includes linear inter-symbol
interference based on a faster-than-Nyquist signaling rate and a
tight frequency roll-off, and non-linear interference based on
high-power amplification for transmission over the wireless
channel. The signal is received over a non-linear wireless channel,
and is processed via a plurality of decoding iterations. A set of
soft information of a current decoding iteration is generated based
on a current estimate of the source signal and a final set of soft
information from a previous decoding iteration. The current
estimate of the source signal is based on an estimate of the linear
ISI and the non-linear interference, which is based on the final
set of soft information from the previous decoding iteration.
Inventors: |
BEIDAS; Bassel; (Alexandria,
VA) ; SESHADRI; Rohit Iyer; (Gaithersburg, MD)
; EROZ; Mustafa; (Germantown, MD) ; LEE;
Lin-Nan; (Potomac, MD) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Hughes Network Systems, LLC |
Germantown |
MD |
US |
|
|
Family ID: |
52132828 |
Appl. No.: |
14/326418 |
Filed: |
July 8, 2014 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
61843905 |
Jul 8, 2013 |
|
|
|
Current U.S.
Class: |
375/341 |
Current CPC
Class: |
H04L 25/03834 20130101;
H04L 25/0314 20130101; H04L 1/0054 20130101; H04L 1/0048 20130101;
H04L 25/497 20130101; H04L 25/03006 20130101; H04L 25/03165
20130101; H04L 1/005 20130101 |
Class at
Publication: |
375/341 |
International
Class: |
H04L 25/03 20060101
H04L025/03; H04L 1/00 20060101 H04L001/00 |
Claims
1. An apparatus comprising: a receiver module configured to process
a signal received over a wireless channel, wherein the received
signal reflects a source signal comprising a plurality of source
symbols, and includes linear inter-symbol interference (ISI)
effects induced based on one or more of a faster-than-Nyquist (FTN)
signaling rate and a tight frequency roll-off applied to the source
signal, and non-linear interference effects induced based on
high-power amplification for transmission over the wireless
channel; wherein the receiver module is configured to process the
received signal based on a plurality of decoding iterations, and
wherein the receiver module comprises: a likelihood metric
computing module configured to generate a set of soft information
of a current decoding iteration based on a current estimate of the
source signal and a final set of soft information from a previous
decoding iteration; wherein the current estimate of the source
signal is based on an estimate of the linear ISI effects and the
non-linear interference effects, which is based on the final set of
soft information from the previous decoding iteration.
2. The apparatus of claim 1, wherein the likelihood metric
computing module comprises a log-likelihood ratio (LLR) computation
module, and the set of soft information of the current decoding
iteration comprises a set of extrinsic log-likelihood ratios (LLRs)
of the current decoding iteration, generated by the LLR computation
module based on the estimate of the source signal and a set of
extrinsic LLRs from the previous decoding iteration.
3. The apparatus of claim 1, wherein the receiver module further
comprises: a mapper module configured to bit-to-symbol map the
final set of soft information from the previous decoding iteration;
a filter module configured to process the bit-to-symbol mapped
information generated by the mapper module to generate the estimate
of the linear ISI effects and the non-linear interference effects;
and an arithmetic module configured to generate the current
estimate of the source signal by subtracting the estimated linear
ISI effects and non-linear interference effects generated by the
filter module from the received signal.
4. The apparatus of claim 3, wherein the filter module comprises a
Volterra filter configured to generate the estimate of the linear
ISI effects and the non-linear interference effects, wherein the
Volterra filter comprises: a first order component used to generate
an estimate of the first order interference, which reflects the
linear ISI effects; and a third order component used to generate an
estimate of the third order interference, which reflects the
non-linear interference effects.
5. The apparatus of claim 1, wherein the final set of soft
information from the previous decoding iteration is generated based
on a plurality of likelihood metric computation iterations,
wherein: for each likelihood metric computation iteration (except
for a first iteration), the likelihood metric computing module is
configured to generate an updated set of soft information based on
a set of soft information of a previous likelihood metric
computation iteration; and the final set of soft information is
based on the updated set of soft information generated by the
likelihood metric computing module as a result of a final
likelihood metric computation iteration.
6. The apparatus of claim 5, wherein the receiver module further
comprises: a decoder module configured to decode, for each
likelihood metric computation iteration, the updated set of soft
information generated by the likelihood metric computing module;
wherein the final set of soft information is based on the decoded
information generated by the decoder module as a result of the
final likelihood metric computation iteration.
7. The apparatus of claim 6, wherein the receiver module further
comprises: a deinterleaver module configured to deinterleave, for
each likelihood metric computation iteration, the updated set of
soft information generated by the likelihood metric computing
module prior to being decoded by the decoder module; and an
interleaver module configured to interleave, for each likelihood
metric computation iteration, the decoded information generated by
the decoder module to generate the set of soft information of the
previous likelihood metric computation iteration; wherein the final
set of soft information is comprised of the interleaved information
generated by the interleaver module as a result of the final
likelihood metric computation iteration.
8. The apparatus of claim 7, wherein the receiver module further
comprises: a mapper module configured to bit-to-symbol map the
final set of soft information; a filter module configured to
process the bit-to-symbol mapped information generated by the
mapper module to generate the estimate of the linear ISI effects
and the non-linear interference effects; and an arithmetic module
configured to generate the current estimate of the source signal by
subtracting the estimated linear ISI effects and non-linear
interference effects generated by the filter module from the
received signal.
9. The apparatus of claim 8, wherein the filter module comprises a
Volterra filter configured to generate the estimate of the linear
ISI effects and the non-linear interference effects, wherein the
Volterra filter comprises: a first order component used to generate
an estimate of the first order interference, which reflects the
linear ISI effects; and a third order component used to generate an
estimate of the third order interference, which reflects the
non-linear interference effects.
10. The apparatus of claim 9, wherein the likelihood metric
computing module comprises a log-likelihood ratio (LLR) computation
module, and the set of soft information of the current decoding
iteration comprises a set of extrinsic log-likelihood ratios (LLRs)
of the current decoding iteration, generated by the LLR computation
module based on the estimate of the source signal and a set of
extrinsic LLRs from the previous decoding iteration.
11. A method comprising: processing a signal received over a
wireless channel, wherein the received signal reflects a source
signal comprising a plurality of source symbols, and includes
linear inter-symbol interference (ISI) effects induced based on one
or more of a faster-than-Nyquist (FTN) signaling rate and a tight
frequency roll-off applied to the source signal, and non-linear
interference effects induced based on high-power amplification for
transmission over the wireless channel; wherein the processing of
the received signal is performed based on a plurality of decoding
iterations, and wherein the processing of the received signal
comprises generating a set of soft information of a current
decoding iteration based on a current estimate of the source signal
and a final set of soft information from a previous decoding
iteration; and wherein the current estimate of the source signal is
based on an estimate of the linear ISI effects and the non-linear
interference effects, which is based on the final set of soft
information from the previous decoding iteration.
12. The method of claim 11, wherein the set of soft information of
the current decoding iteration comprises a set of extrinsic
log-likelihood ratios (LLRs) of the current decoding iteration,
generated based on the estimate of the source signal and a set of
extrinsic LLRs from the previous decoding iteration.
13. The method of claim 11, wherein the processing of the received
signal further comprises: bit-to-symbol mapping the final set of
soft information from the previous decoding iteration; filtering
the bit-to-symbol mapped information to generate the estimate of
the linear ISI effects and the non-linear interference effects; and
generating the current estimate of the source signal by subtracting
the estimated linear ISI effects and non-linear interference
effects from the received signal.
14. The method of claim 13, wherein the filtering of the
bit-to-symbol mapped information is based on a Volterra filter
model configured to generate the estimate of the linear ISI effects
and the non-linear interference effects, wherein the Volterra
filter model comprises: a first order component used to generate an
estimate of the first order interference, which reflects the linear
ISI effects; and a third order component used to generate an
estimate of the third order interference, which reflects the
non-linear interference effects.
15. The method of claim 11, wherein the final set of soft
information from the previous decoding iteration is generated based
on a plurality of likelihood metric computation iterations,
wherein: for each likelihood metric computation iteration (except
for a first iteration), an updated set of soft information is
generated based on a set of soft information of a previous
likelihood metric computation iteration; and the final set of soft
information is based on the updated set of soft information
generated as a result of a final likelihood metric computation
iteration.
16. The method of claim 15, wherein the processing of the received
signal further comprises: decoding, for each likelihood metric
computation iteration, the updated set of soft information; wherein
the final set of soft information is based on the decoded
information generated as a result of the final likelihood metric
computation iteration.
17. The method of claim 16, wherein the processing of the received
signal further comprises: deinterleaving, for each likelihood
metric computation iteration, the updated set of soft information
prior to the decoding; and interleaving, for each likelihood metric
computation iteration, the decoded information to generate the set
of soft information of the previous likelihood metric computation
iteration; and wherein the final set of soft information is
comprised of the interleaved information generated as a result of
the final likelihood metric computation iteration.
18. The method of claim 17, wherein the processing of the received
signal further comprises: bit-to-symbol mapping the final set of
soft information; filtering the bit-to-symbol mapped information to
generate the estimate of the linear ISI effects and the non-linear
interference effects; and generating the current estimate of the
source signal by subtracting the estimated linear ISI effects and
non-linear interference effects from the received signal.
19. The method of claim 18, wherein the filtering of the
bit-to-symbol mapped information is based on a Volterra filter
model configured to generate the estimate of the linear ISI effects
and the non-linear interference effects, wherein the Volterra
filter model comprises: a first order component used to generate an
estimate of the first order interference, which reflects the linear
ISI effects; and a third order component used to generate an
estimate of the third order interference, which reflects the
non-linear interference effects.
20. The method of claim 19, wherein the set of soft information of
the current decoding iteration comprises a set of extrinsic
log-likelihood ratios (LLRs) of the current decoding iteration,
generated based on the estimate of the source signal and a set of
extrinsic LLRs from the previous decoding iteration.
Description
RELATED APPLICATIONS
[0001] This application claims the benefit of the earlier filing
date under 35 U.S.C. .sctn.119(e) of U.S. Provisional Application
Ser. No. 61/843,905 (filed 2013-07-08).
BACKGROUND
[0002] The present invention generally covers receivers in wireless
communications systems, and more specifically is generally drawn to
addressing noise and/or interference effects exhibited by received
signals, where the signals were transmitted via a transmitter
employing high power amplifiers (HPAs), such as satellite
transponders in a satellite communications system. A satellite
communication system may include a transmitter having a high power
amplifier (HPA) or a transponder that includes a transmitter having
an HPA. The output of a transmitter can be seen as a sequence of
symbols called a phrase. Each symbol represents a sequence of bits
(e.g., in the case of 8PSK, each symbol represents 3 bits), and the
transmitter will output the phrase one symbol at a time during
transmission. As a transmitter shifts from one symbol to the next
in the phrase, previous output symbols may cause interference in
the output of the current symbol. Similarly, the current symbol is
also affected by interference resulting from subsequent or future
symbols. This interference in the current symbol caused by previous
as well as symbols is referred to as the inter-symbol interference
(ISI). ISI represents a form of signal distortion whereby one
symbol interferes with subsequent symbols. ISI is usually caused by
multipath propagation, or the inherent non-linear frequency
response of a channel causing successive symbols to blur together.
Further, typically, an HPA operates most efficiently at or near
saturation, however, operation of an HPA at or near saturation
generates nonlinear distortion in output channels. ISI can be
mitigated by reducing the transmission or throughput rate of the
transmitter, however, a reduction in the throughput rate
proportionately reduces bandwidth efficiency.
[0003] In order to increase system throughput, a logical goal would
be to maximize the number of transponders/HPAs of the satellite
transmission antenna. Due to physical limitations, however, there
is a maximum number of HPA units that can fit in a single
transponder. To combat this issue, multiple carriers can be shared
by a single transponder HPA (multicarrier operation), allowing for
the transmission of more data and the servicing more users without
exceeding the physical limitation on the number of HPAs per
transponder. Another benefit of multicarrier operation is that it
facilitates a reduction of the transmission symbol rate per carrier
without sacrificing system throughput, which greatly eases the
burden on hardware implementation. In a multicarrier system,
however, the amplification of multiple carriers by way of a single
HPA (driven at or near its saturation point for maximum efficiency)
generates a large amount of nonlinear interference or distortion,
which further contributes to performance degradation issues.
[0004] Additionally, in order to increase transmission throughput,
the transmission rate or symbol rate (in the time domain) can be
increased, without altering the spectral shape of the signal.
Increasing the transmission throughput, however, further
exacerbates ISI issues. According to the Nyquist theorem, there is
an ideal transmission limit (the Nyquist rate) beyond which the
ambiguity in ability to resolve symbols at the receiver
increases--the maximum number of code elements per second that
could be unambiguously resolved at the receiver. Transmission at
the Nyquist rate mitigates ISI, while increasing the transmission
throughput above the Nyquist rate (at a "faster than Nyquist (FTN)"
rate) resulting in linear interference that exacerbates the issues
of ISI.
[0005] Further, in order to increase spectral efficiency, it is
desirable to pack channels closer together in the frequency domain,
which results in increased throughput (e.g., in bits/second/Hz,
where the Hz reflects the distance between adjacent channels). The
spectral efficiency, however, is constrained by the roll-off
factor, which reflects the rate of slope or steepness of a
transmission function with respect to frequency. The slower the
roll-off rate (or the higher the roll-off percentage or factor) the
further apart the adjacent channels must be placed to mitigate
adjacent channel interference (ACI). ACI results from extraneous
power picked up from a signal in an adjacent channel (e.g., one
channel bleeds-over into an adjacent channel). Accordingly, the
slower the roll-off rate of a channel, the higher the signal power
that can be picked up by an adjacent channel. Therefore, there is
an inherent tradeoff between roll-off rate and spectral
efficiency.
[0006] Accordingly, to maximize bandwidth efficiency of a system,
two goals are to increase transmission throughput of a transponder
(transmission rate) in the time domain, and to increase the rate or
steepness of the roll-off (operate at a decreased or minimized
roll-off factor or percentage). As described above, however, an
increase in the transmission throughput beyond certain levels and
tightening the roll-off contributes to both ISI and ACI. More
specifically, the resulting interference manifests itself as a
structured interference, which is significant and extends for a
relatively longer period in the time domain (the interference tends
to linger in time over many symbols, resulting in a significant
degradation in performance). At the receiver, in view of the
lengthened period of significant interference, the receiver must be
configured to handle the increased interference levels, which would
require increased complexity in the receiver. The longer the
interference memory, the receiver must account for the possible
sequences, which is exponential in the symbol alphabet over that
memory. For example, with a 16APSK modulation scheme, the receiver
would be required to consider 16 raised to the power of the channel
interference memory signal possibilities in the decoding process.
In other words, the receiver must be configured to account for a
significantly increased number of possibilities for the transmitted
signal before making a decoder decision.
[0007] Further, due to physical limitations of the satellite, there
are a maximum number of HPA units that can fit in a transponder. To
solve the issue of such physical limitations, sharing multiple
carriers by a single transponder HPA (multicarrier operation)
allows for transmitting more data and servicing more users. Another
benefit of multicarrier operation is that it allows for reducing
the transmission symbol rate per carrier without sacrificing system
throughput. This greatly eases the burden on hardware
implementation. When multiple carriers are amplified by way of a
single HPA, and when the HPA is driven near its saturation point, a
significant level of nonlinear interference is generated.
Interference is an undesirable result of increasingly crowded
spectrum, when multiple carriers share the same transponder high
power amplifier (HPA). The transponder HPA transmits a maximum
signal strength when operating at or near its saturation output
power level. Operating near saturation, however, increases
nonlinearities in the HPA, and such nonlinearities in the HPA
result in nonlinear distortion (e.g., intermodulation distortion
(IMD), which comprises unwanted amplitude and phase modulation of
signals containing two or more different frequencies in a system
with nonlinearities). The intermodulation between each frequency
component will form additional signals at frequencies that are not,
in general, at harmonic frequencies (integer multiples) of either,
but instead often at sum and difference frequencies of the original
frequencies. The spurious signals, which are generated due to the
nonlinearity of a system, are mathematically related to the
original input signals. When the spurious signals are of sufficient
amplitude, they can cause interference within the original system
or in other systems, and, in extreme cases, loss of transmitted
information, such as voice, data or video.
[0008] IMD causes interference within a message itself as well as
between the message signals by transferring modulations from one
frequency range to another. The problem is particularly acute when
a cost effective nonlinearized HPA is operated with minimal output
back-off (OBO). OBO is the amount (in dB) by which the output power
level of the HPA is reduced, or "backed-off," from the saturation
output power level. The problem is further compounded when the
carriers passing through the HPA are bandwidth efficient, whose
constellations include multiple concentric rings, and the carriers
are tightly spaced within the limited spectrum. The interference
issues are further complicated when transmission throughput of a
transponder (the symbol transmission rate) is increased in the time
domain (e.g., an FTN rate) and the rate or steepness of the
roll-off is increased. As described above, however, an increase in
the transmission throughput beyond certain levels (e.g., the
Nyquist level) and tightening the roll-off contributes to both ISI
and ACI.
[0009] Band-pass filtering can be an effective way to eliminate
most of the undesired products without affecting in-band
performance. However, third order intermodulation products are
usually too close to the fundamental signals and cannot be easily
filtered. The amplitude and phase distortion is unacceptable in
systems that use higher order modulation schemes, because the
distortion results in an error component in the received vector,
degrading the receiver's bit error rate (BER). Other attempts to
compensate for nonlinear interference have been complex and require
receivers to exchange information. For instance, a conventional
system compensates for linear and nonlinear ISI) and linear and
nonlinear adjacent channel interference (ACI) due to the
nonlinearlity of HPA and tight crowding of carriers in a
transmitter HPA or transmitter section of a transponder HPA.
However, such a system requires receivers to coordinate samples
from adjacent carriers, resulting in increased system complexity
and computational effort.
[0010] What is needed, therefore, is an approach for increasing the
transmission throughput of a wireless transmitter or transponder
HPA driven at or near saturation, while being able to efficiently
decode the transmitted signal at a receiver.
Some Example Embodiments
[0011] The present invention advantageously addresses the needs
above, as well as other needs, by providing a system that applies
faster-than-Nyquist "FTN" transmission symbol rates, combined with
tight frequency roll-off, and employs a receiver that includes
novel interference compensation techniques (capable of handling the
enhanced level of non-linear distortion or interference
attributable to the HPAs of the satellite transponders, and the
linear interference resulting from the IMUX and OMUX filters of the
satellite transponders, and the linear interference or enhanced ISI
memory attributable to the FTN transmission rates), while
maintaining a complexity that does not grow exponentially with the
interference memory and signal constellation size.
[0012] In accordance with example embodiments, an apparatus
comprises a receiver module configured to process a signal received
over a wireless channel, wherein the received signal reflects a
source signal comprising a plurality of source symbols, and
includes linear inter-symbol interference (ISI) effects induced
based on one or more of a faster-than-Nyquist (FTN) signaling rate
and a tight frequency roll-off applied to the source signal, and
non-linear interference effects induced based on high-power
amplification for transmission over the wireless channel. The
receiver module is configured to process the received signal based
on a plurality of decoding iterations, and wherein the receiver
module comprises a likelihood metric computing module configured to
generate a set of soft information of a current decoding iteration
based on a current estimate of the source signal and a final set of
soft information from a previous decoding iteration. The current
estimate of the source signal is based on an estimate of the linear
ISI effects and the non-linear interference effects, which is based
on the final set of soft information from the previous decoding
iteration. By way of example, the likelihood metric computing
module comprises a log-likelihood ratio (LLR) computation module,
and the set of soft information of the current decoding iteration
comprises a set of extrinsic log-likelihood ratios (LLRs) of the
current decoding iteration, generated by the LLR computation module
based on the estimate of the source signal and a set of extrinsic
LLRs from the previous decoding iteration.
[0013] According to one embodiment of the apparatus, the final set
of soft information from the previous decoding iteration is
generated based on a plurality of likelihood metric computation
iterations. For each likelihood metric computation iteration
(except for a first iteration), the likelihood metric computing
module is configured to generate an updated set of soft information
based on a set of soft information of a previous likelihood metric
computation iteration. The final set of soft information is based
on the updated set of soft information generated by the likelihood
metric computing module as a result of a final likelihood metric
computation iteration.
[0014] According to further embodiments, the receiver module may
further comprise a decoder module configured to decode, for each
likelihood metric computation iteration, the updated set of soft
information generated by the likelihood metric computing module,
wherein the final set of soft information is based on the decoded
information generated by the decoder module as a result of the
final likelihood metric computation iteration. The receiver module
may further comprise a deinterleaver module configured to
deinterleave, for each likelihood metric computation iteration, the
updated set of soft information generated by the likelihood metric
computing module prior to being decoded by the decoder module, and
an interleaver module configured to interleave, for each likelihood
metric computation iteration, the decoded information generated by
the decoder module to generate the set of soft information of the
previous likelihood metric computation iteration. According to such
an embodiment, the final set of soft information is comprised of
the interleaved information generated by the interleaver module as
a result of the final likelihood metric computation iteration. The
receiver module may further comprise a mapper module configured to
bit-to-symbol map the final set of soft information, a filter
module configured to process the bit-to-symbol mapped information
generated by the mapper module to generate the estimate of the
linear ISI effects and the non-linear interference effects, and an
arithmetic module configured to generate the current estimate of
the source signal by subtracting the estimated linear ISI effects
and non-linear interference effects generated by the filter module
from the received signal. By way of example, the filter module may
comprise a Volterra filter configured to generate the estimate of
the linear ISI effects and the non-linear interference effects,
wherein the Volterra filter comprises a first order component used
to generate an estimate of the first order interference, which
reflects the linear ISI effects, and a third order component used
to generate an estimate of the third order interference, which
reflects the non-linear interference effects.
[0015] In accordance with further example embodiments, a method
comprises processing a signal received over a wireless channel,
wherein the received signal reflects a source signal comprising a
plurality of source symbols, and includes linear inter-symbol
interference (ISI) effects induced based on one or more of a
faster-than-Nyquist (FTN) signaling rate and a tight frequency
roll-off applied to the source signal, and non-linear interference
effects induced based on high-power amplification for transmission
over the wireless channel. The processing of the received signal is
performed based on a plurality of decoding iterations, wherein the
processing of the received signal comprises generating a set of
soft information of a current decoding iteration based on a current
estimate of the source signal and a final set of soft information
from a previous decoding iteration. The current estimate of the
source signal is based on an estimate of the linear ISI effects and
the non-linear interference effects, which is based on the final
set of soft information from the previous decoding iteration. By
way of example, the set of soft information of the current decoding
iteration comprises a set of extrinsic log-likelihood ratios (LLRs)
of the current decoding iteration, generated based on the estimate
of the source signal and a set of extrinsic LLRs from the previous
decoding iteration.
[0016] According to one embodiment of the method, the final set of
soft information from the previous decoding iteration is generated
based on a plurality of likelihood metric computation iterations.
For each likelihood metric computation iteration (except for a
first iteration), an updated set of soft information is generated
based on a set of soft information of a previous likelihood metric
computation iteration, and the final set of soft information is
based on the updated set of soft information generated as a result
of a final likelihood metric computation iteration.
[0017] According to further embodiments, the processing of the
received signal may further comprise decoding, for each likelihood
metric computation iteration, the updated set of soft information,
wherein the final set of soft information is based on the decoded
information generated as a result of the final likelihood metric
computation iteration. The processing of the received signal may
further comprise deinterleaving, for each likelihood metric
computation iteration, the updated set of soft information prior to
the decoding, and interleaving, for each likelihood metric
computation iteration, the decoded information to generate the set
of soft information of the previous likelihood metric computation
iteration. According to such an embodiment, the final set of soft
information is comprised of the interleaved information generated
as a result of the final likelihood metric computation iteration.
The processing of the received signal may further comprise
bit-to-symbol mapping the final set of soft information, filtering
the bit-to-symbol mapped information to generate the estimate of
the linear ISI effects and the non-linear interference effects, and
generating the current estimate of the source signal by subtracting
the estimated linear ISI effects and non-linear interference
effects from the received signal. By way of example, the filtering
of the bit-to-symbol mapped information may be based on a Volterra
filter model configured to generate the estimate of the linear ISI
effects and the non-linear interference effects, wherein the
Volterra filter model comprises a first order component used to
generate an estimate of the first order interference, which
reflects the linear ISI effects, and a third order component used
to generate an estimate of the third order interference, which
reflects the non-linear interference effects.
[0018] Still other aspects, features, and advantages of the present
invention are readily apparent from the following detailed
description, simply by illustrating a number of particular
embodiments and implementations, including the best mode
contemplated for carrying out the present invention. The present
invention is also capable of other and different embodiments, and
its several details can be modified in various obvious respects,
all without departing from the spirit and scope of the present
invention. Accordingly, the drawing and description are to be
regarded as illustrative in nature, and not as restrictive.
BRIEF DESCRIPTION OF THE DRAWINGS
[0019] The present invention is illustrated by way of example, and
not by way of limitation, in the figures of the accompanying
drawings and in which like reference numerals refer to similar
elements and in which:
[0020] FIGS. 1A and 1B illustrate communications systems capable of
employing an interference compensation system and algorithms, in
accordance with example embodiments of the present invention;
[0021] FIG. 2 illustrates a block diagram depicting an example
transmitter and receiver of the communications system of FIGS. 1A
and 1B, in accordance with example embodiments of the present
invention;
[0022] FIG. 3 illustrates the ISI introduced in the case of sharp
spectral roll-off (where the roll-off is 5% and the FTN rate is
25%), in accordance with example embodiments of the present
invention;
[0023] FIGS. 4A and 4B illustrate example amplitude and group delay
responses of a typical input multiplexer (IMUX) of the satellite of
FIG. 1B, in accordance with example embodiments of the present
invention;
[0024] FIGS. 5A and 5B illustrate example non-linearized AM/AM and
AM/PM distortion characteristics of a typical traveling wave tube
amplifier (TWTA) of the satellite of FIG. 1B, in accordance with
example embodiments of the present invention;
[0025] FIGS. 6A and 6B illustrate example amplitude and group delay
responses of a typical output multiplexer (OMUX) of the satellite
of FIG. 1B, in accordance with example embodiments of the present
invention;
[0026] FIG. 7 illustrates a block diagram of the Turbo Volterra
Module of the receivers of FIG. 2, in accordance with example
embodiments of the present invention;
[0027] FIG. 8 illustrates packet error rate (PER) performance
curves (as a function of the per-symbol signal-to-noise ratio
(SNR)) for an example system employing IMUX and MUX filters and a
TWTA high power amplifier within the transponders of the satellite,
wherein, in each case, the system achieves a spectral efficiency of
2.42 bps/Hz, and each curve reflects application of a particular
modulation and coding scheme in combination with either a prior art
least mean square (LMS) adaptive equalizer or a Turbo Volterra
Module (of example embodiments of the present invention) employed
within the receiver;
[0028] FIG. 9 illustrates a computer system upon which example
embodiments according to the present invention can be implemented;
and
[0029] FIG. 10 is a diagram of a chip set that can be utilized in
implementing an interference compensation system, according to
example embodiments.
DETAILED DESCRIPTION
[0030] A system that applies faster-than-Nyquist "FTN" transmission
symbol rates, combined with tight frequency roll-off, and employs a
receiver that includes novel interference compensation techniques
(capable of handling the enhanced level of non-linear distortion or
interference attributable to the HPAs of the satellite
transponders, and the linear interference resulting from the IMUX
and OMUX filters of the satellite transponders, and the linear
interference or enhanced ISI memory attributable to the FTN
transmission rates), is described. In the following description,
for the purposes of explanation, numerous specific details are set
forth in order to provide a thorough understanding of the
invention. It is apparent, however, that the invention may be
practiced without these specific details or with an equivalent
arrangement. In other instances, well known structures and devices
are shown in block diagram form in order to avoid unnecessarily
obscuring the invention.
[0031] FIG. 1A illustrates a block diagram of a communications
system capable of employing an interference compensation system and
algorithms, in accordance with example embodiments of the present
invention. With reference to FIG. 1A, a broadband communications
system 110 includes one or more transmitters 112 (of which one is
shown) that generate signal waveforms across a communications
channel 114 to one or more receivers 116 (of which one is shown).
In this discrete communications system 110, the transmitter 112 has
a signal source that produces a discrete set of data signals, where
each of the data signals has a corresponding signal waveform. These
signal waveforms are attenuated, or otherwise altered, by
communications channel 114. Coding may be utilized to combat noise
and other issues associated with the channel 114, such as forward
error correction (FEC) codes.
[0032] FIG. 1B illustrates an example satellite communications
system 130 capable of supporting communications among terminals
with varied capabilities, including an interference compensation
system and algorithms, in accordance with example embodiments of
the present invention. Satellite communications system 130 includes
a satellite 132 that supports communications among multiple
satellite terminals (STs) 134a-134n, a number of gateways (GWs)
138a-138n, and a network operations center (NOC) 142. The STs, GWs
and NOC transmit and receive signals via the antennas 136a-136n,
146a-146n, and 156, respectively. According to different
embodiments, the NOC 142 may reside at a separate site reachable
via a separate satellite channel or may reside within a GW site.
The NOC 142 performs the management plane functions of the system
130, while the GWs 138a-138n perform the data plane functions of
the system 133. For example, the NOC 142 performs such functions as
network management and configuration, software downloads (e.g., to
the STs 134a-134n), status monitoring, statistics functions (e.g.,
collection, aggregation and reporting), security functions (e.g.,
key generation, management and distribution), ST registration and
authentication, and GW diversity management. The NOC 142
communicates with each GW via the satellite 132, or via a secure
private communications network 152 (e.g., an IPsec tunnel over a
dedicated link or a virtual private network (VPN) or IPsec tunnel
through a public network, such as the Internet). It should be noted
that, according to one example embodiment, the traffic
classification approaches of embodiments of the present invention
address classification of data traffic flowing through an
aggregation point or node. Additionally, each GW and the NOC have
connectivity to one or more public communications networks, such as
the Internet or a PSTN.
[0033] According to a further example embodiment, each of the GWs
138a-138n include one or more IP gateways (IPGWs)--whereby the data
plane functions are divided between a GW and its respective IPGWs.
For example, GW 138a includes IPGWs 148a(1)-148a(n) and GW 138n
includes IPGWs 148n(1)-148n(n). A GW may perform such functions as
link layer and physical layer outroute coding and modulation (e.g.,
DVB-S2 adaptive coding and modulation), link layer and physical
layer inroute handling (e.g., IPOS), inroute bandwidth allocation
and load balancing, outroute prioritization, web acceleration and
HTTP compression, flow control, encryption, redundancy switchovers,
and traffic restriction policy enforcement. Whereas, the IPGW may
perform such functions as data compression, TCP performance
enhancements (e.g., TCP performance enhancing proxies, such as TCP
spoofing), quality of service functions (e.g., classification,
prioritization, differentiation, random early detection (RED),
TCP/UDP flow control), bandwidth usage policing, dynamic load
balancing, and routing. Further, a GW and respective IPGW may be
collocated with the NOC 142. The STs 134a-134n provide connectivity
to one or more hosts 144a-144n and/or routers 154a-154n,
respectively. The Satellite communications system 130 may operate
as a bent-pipe system, where the satellite essentially operates as
a repeater or bent pipe. Alternatively, the system 130 may employ a
switching or processing satellite supporting mesh communications
(point-to-point communications directly between, for example, the
two STs 134a and 134n).
[0034] In a bent-pipe system of an example embodiment, the
satellite 132 operates as a repeater or bent pipe, and
communications to and from the STs 134a-134n are transmitted over
the satellite 132 to and from respective IPGWs associated with
particular STs. Further, in a spot beam system, any one spot beam
operates as a bent-pipe to geographic region covered by the beam.
For example, each spot beam operates as a bent pipe communications
channel to and from the STs and/or IPGW(s) within the geographic
region covered by the beam. Accordingly, signal transmissions to
the satellite are either from an ST and destined for an associated
gateway, or from a gateway and destined for an associated ST.
According to one embodiment, several GWs/IPGWs are distributed
across the geographic region covered by all spot beams of the
satellite 132, where, in a beam in which a GW (and respective
IPGWs) are located, only the one GW (and no STs) occupies that
beam. Further, each IPGW may serve as an aggregation node for a
multitude of remote nodes or STs. The total number of GWs/IPGWs,
and the geographic distribution of the GWs/IPGWs, depends on a
number of factors, such as the total capacity of the satellite
dedicated to data traffic, geographic traffic loading of the system
(e.g., based on population densities and the geographic
distribution of the STs), locations of available terrestrial data
centers (e.g., terrestrial data trunks for access to public and
private dedicated networks).
[0035] FIG. 2 illustrates a block diagram depicting an example
transmitter and receiver of the communications system of FIGS. 1A
and 1B, where the receiver employs a Turbo Volterra Module for
interference compensation and algorithms, in accordance with
example embodiments of the present invention. While embodiments of
the present invention are not limited to a satellite communications
system, for the purpose of explanation, the following description
envisions an embodiment encompassing the satellite communications
system 130 of FIG. 1B. As illustrated in FIG. 2, the communication
system includes transmitters 201 (201a-201m) and receivers 203
(203a-203m), with the signals being transmitted over the channel
114, via the transponder/amplifier 217, where the
transponder/amplifier 217 comprises components of the transmission
section of the satellite 132. The transmitters 201a-201m and
receivers 203a-203m may represent a corresponding number of STs 134
and GWs 138. By way of example, a particular transmission 223a may
reflect a transmission of data from a data source 205a (e.g., the
host 144a), by the ST 134a, and destined for the GW 138a, where the
receiver portion of the GW 138a may comprise the receiver 203a. A
transmitter 201, in accordance with example embodiments, generally
comprises at least one data or signal source 205, an encoder
section 207, a modulator section 209, a filter section 211 and a
transmitter section 213 (e.g., an upconverter/amplifier section). A
receiver 203, in accordance with example embodiments, generally
comprises a receiver section 231, a filter section 232, an
IMUX/OMUX equalizer section 233, a sampler module 239, and a Turbo
Volterra Module 235 (which includes de-interleaver and decoder
sections, as depicted in FIG. 7).
[0036] According to one example embodiment the satellite system
comprises a bent-pipe system, where the satellite acts as a
repeater (as described above). The transponder of such a
communications satellite comprises a series of interconnected
components that form a communications channel between the satellite
receive and transmit antennas. At the receive side, a typical
transponder generally comprises an input band limiting device
(e.g., a band pass filter), an input low-noise amplifier (LNA)
(which amplifies the received signal to compensate for the
significant weakening of the signal due to large distance traveled
between the earth station transmitter and the satellite), and an
input multiplexer (IMUX) (which generally comprises filter banks
that channelize the receive band into the individual channels). At
the transmit side, a typical transponder generally comprises a
frequency translator (which converts the frequency of the received
signal to the frequency required for the transmitted signal), an
output band-limiting device (e.g., a band=pass filter), and a
downlink high power amplifier (HPA) (which amplifies the signal for
transmission back down to an earth station receiver). In one
embodiment, due to the physical limitations of the number of HPAs
that can fit in the downlink transmission section of the satellite
132, to maximize bandwidth efficiencies (e.g., to increase
bandwidth and data throughput), multiple received uplink channels
or carrier signals can be multiplexed onto a single wideband
carrier of a single downlink transponder HPA 217 (a wideband
multi-carrier system). In such a multicarrier system, the downlink
transponder will also include a signal combiner section or output
multiplexer (OMUX), which combines the uplink transponder channels
or carrier signals that are switched for transmission to a common
downlink cell 230. The OMUX thereby generates a combined
transmission signal for transmission via the HPA for the particular
transmit signal or downlink beam 225.
[0037] Accordingly, in such a multi-carrier system, the satellite
aggregates multiple received uplink data signals (e.g., data
signals destined for a particular geographic region serviced by a
particular downlink beam of the satellite), where each uplink data
signal is carried by a separate carrier. The satellite
simultaneously transmits the aggregate data signal over the single
downlink channel 227 to the single downlink cell 230, which is
transmitted via a single downlink transponder HPA 217, on a single
downlink signal 225. During transmission over the downlink channel
227, the transmitted downlink signal 225 will encounter various
physical effects that manifest as noise experienced in the received
signal. The added channel noise typically may be idealized as
additive white Gaussian noise. Hence, the transmitted signal 225
reflects multiple source data signals 223a-223m, respectively
carrying data generated by the different data sources 205a-205m.
While a variable number of data signals may be transmitted over the
satellite 132 via such a multi-carrier system, however, for
purposes of simplification, the following description envisions an
embodiment encompassing data signals from two signal sources 205a
and 205m, respectively transmitted via the uplink transmission
signals 223a and 223m, and combined via the satellite and
transmitted back via the downlink transmission signal 225.
[0038] According to an example embodiment, in operation, data or
signal source 205a outputs a first source signal to encoder 207a,
where the first source signal reflects a sequence of source data
symbols for transmission over the communications system. Encoder
207a generates an encoded vector signal b.sub.1 from the first
source signal. In one embodiment, encoder 207a is an error
correction encoder that adds information to reduce information loss
at the receive section 203. Additionally, or alternatively, the
encoder 207a interleaves data from the first source signal into the
encoded vector signal. Modulator 209a receives the encoded vector
signal and generates a modulated discrete signal a.sub.1(t), where
each source symbol is mapped to a respective signal constellation
point of the signal constellation of the applied modulation scheme.
In one embodiment, modulators 209 are Gray-coded Quadrature
Amplitude Modulation (QAM) modulators or Amplitude and Phase Shift
Keyed (APSK) modulators (e.g., QPSK, 8PSK, 16APSK or 32APSK
modulators). Accordingly, depending on the applied modulation
scheme, each source symbol represents a number of source data bits,
where (via the applied modulation) each source symbol is mapped to
an associated signal constellation point and transmitted to the
satellite via a common uplink transmission carrier. For example,
with 16APSK modulation, each of the 16 constellation points
represents or corresponds to an arrangement of four source data
bits (e.g., 0000, 0001, 0010, . . . , 1111), and (via the applied
modulation) each received data symbol is mapped to its
corresponding or associated constellation point. In one embodiment,
the discrete signal output of the modulator 209 (e.g., the
modulated signal) may be represented as:
a m ( t ) = k = - .infin. .infin. a m , k .delta. ( t - k T s - m T
s ) ( 1 ) ##EQU00001##
[0039] where {a.sub.m,k; m=1, . . . , M.sub.c} are sets of complex
valued data symbols, .delta.(t) is the Dirac delta function, and
.epsilon..sub.m represents the normalized difference in signal
arrival times.
[0040] Filter 211a receives the modulated discrete signal
a.sub.1(t) and generates a continuous filtered signal s.sub.1(t)
reflecting the data of the modulated discrete signal. In one
embodiment, filter 211a is a pulse shaping filter with impulse
responses P.sub.m,T(.tau.) to generate the signal s.sub.m(t)
as:
s.sub.m(t)=.intg..sub.-.infin..sup..infin.a.sub.m(t-.tau.)P.sub.m,T(.tau-
.)d.tau. (2)
Alternatively, in the discrete representation:
s m ( t ) = k = - .infin. .infin. a m , k * p m , T ( t - k .tau. T
s ) , .tau. .ltoreq. 1 ( 3 ) ##EQU00002##
where {a.sub.m,k; m=1, . . . , M.sub.c} are sets of complex-valued
data symbols, M.sub.c represents the number of carriers, and
p.sub.m,T(t) are impulse responses of the pulse shaping
filters.
[0041] In the foregoing signal representations for s.sub.m(t),
1/.tau. is the transmission throughput rate. In traditional
communications systems (based on the Nyquist theorem) the rate
1/.tau. is set at or below unity, which avoids ISI for pulses that
are orthogonal to integral shifts of T.sub.s. According to example
embodiments of the present invention, however, to increase the
system throughput rate, the transmission symbol rate is set at a
faster than Nyquist rate (FTN rate), wherein the rate of 1/.tau. is
configured to be greater than unity. Such rates result in linear
interference (e.g., structured ISI) that needs to be mitigated at
the receiver. Further, the FTN-induced ISI has a memory span that
increases with sharper spectral roll-off and more aggressive FTN
rates. FIG. 3, for example, illustrates the ISI introduced in the
case of sharp spectral roll-off (where the roll-off is 5% and the
FTN rate is 25%), in accordance with example embodiments of the
present invention. As shown in FIG. 3, the ISI introduced by time
packing (FTN rates) for spectrally efficient signals decays at a
low rate, spanning as many as 15 symbols on either side. Mitigating
this type of ISI using receivers of exponential complexity in terms
of signal constellation size and ISI memory length would be
prohibitively complex. Alternatively, such FTN and roll-off induced
ISI can be efficiently mitigated using receivers whose complexity
is non-exponential in terms of signal constellation size and ISI
memory length, in accordance with example embodiments of the
present invention.
[0042] According to one embodiment, the filter P.sub.m,T(.tau.) may
model the cascade of pulse-shaping filters and the on-board input
multiplexing filter (IMUX) of the satellite 132. The individual
signals s.sub.m(t) are then frequency-translated to their
respective slot or center frequency. By way of example, to generate
the first continuous carrier signal, the transmitter section 213a
mixes the continuous filtered signal from the filter 211a with a
local oscillator signal to generate the desired carrier signal,
where the oscillator signal may be represented as
exp(j(2.pi.f.sub.1t+.theta..sub.1))/ {square root over (M.sub.c)},
where f.sub.1 and .theta..sub.1 represent the center frequency and
carrier phase of the first carrier signal, to generate a first
continuous carrier signal 223a for transmission over the uplink
channel to the satellite 132. Additional continuous carrier
signals, for transmitting the data source signals (e.g., from data
source 205m), may be generated using similar processes, where each
additional continuous carrier signal would be of a different center
frequency and carrier phase (e.g., for source 205m, f.sub.m and
.theta..sub.m).
[0043] Each of the discrete signals (e.g., 223a through 223m) are
transmitted to the satellite 132 by the associated transmitter
terminals (e.g., 201a through 201m, respectively) via respective
carriers at different carrier frequencies. Once received by the
satellite 132, based on the respective carrier frequency of each of
the received signals and the destination downlink cell 230, the
satellite forms a composite signal for transmission via a
respective transponder to the destination downlink cell. The
composite signal can be represented in complex form as:
s c ( t ) = m = 1 M c s m ( t ) exp ( j ( 2 .pi. f m t + .theta. m
) ) M c ( 4 ) ##EQU00003##
where f.sub.m and .theta..sub.m are the center frequency and
carrier phase of the m.sup.th uplink channel, respectively. The
composite signal is then processed via the respective satellite
downlink transponder within the transmit section of the satellite
132 (e.g., downlink transponder 217). Within the downlink
transponder 217, the composite signal is processed through an IMUX
filter (e.g., to select the desired carrier and remove any adjacent
carriers). In other words, in a multicarrier system, the IMUX
selects the desired channel/carrier and filters out the other
channels/carriers, and the desired carrier then passes through the
HPA individually. The IMUX thereby tunes the transponder to the
desired carrier frequency for the transmission channel. The IMUX,
however, produces amplitude distortion and group delay and the
group delay causes linear ISI. FIGS. 4A and 4B illustrate example
amplitude and group delay responses of a typical input multiplexer
(IMUX) of the satellite 132, in accordance with example embodiments
of the present invention.
[0044] Further, downlink transponder 217 amplifies the composite
signal (e.g., via an HPA) to generate the downlink transmission
signal 225, which is transmitted to the respective downlink cell
230. In one embodiment, to achieve a maximum efficiency of the
downlink transponder 217 (e.g., to achieve a maximum output power
without overly distorting the amplified signal, and thereby achieve
power and bandwidth or data throughput efficiencies), the HPA is
driven near or to its saturation level, while the back-off is
minimized. The HPA thereby operates in the nonlinear region of its
output range, and, in view of the multiple uplink signals being
transmitted simultaneously, the uplink signal carriers interact
with or affect each other in a nonlinear fashion. Additionally, to
achieve further efficiency, the system may be designed such that a
single downlink HPA 217 may be transmitting signals of differing
rates, employing multiple rate constellations (e.g., QPSK, 8PSK,
16APSK, 32APSK, etc.). By way of example, the HPA may comprise a
traveling wave tube amplifier (TWTA)) operating at an optimized
back-off level (e.g., driven at or near saturation). FIGS. 5A and
5B illustrate example non-linearized AM/AM and AM/PM distortion
characteristics of a typical traveling wave tube amplifier (TWTA)
of the satellite 132, in accordance with example embodiments of the
present invention. These graphs illustrate the non-linear
distortion effects of such amplifiers at different back-off levels.
Accordingly, the HPA introduces a significant level of distortion
(e.g., nonlinear interference) resulting in significant nonlinear
ISI in the transmitted signal 225. The amplified signal output from
the HPA is then fed through an output multiplexer (OMUX) within the
downlink transponder 217. The OMUX filter is applied to the
amplified signal to limit the interference to adjacent
transponders. As with the IMUX, however, the OMUX also produces
amplitude distortion and group delay and the group delay again
causes linear ISI. FIGS. 6A and 6B illustrate example amplitude and
group delay responses of a typical output multiplexer (OMUX), in
accordance with example embodiments of the present invention.
[0045] The receivers 203 located within the downlink cell 230, that
is serviced by the downlink beam or channel 227, all receive the
same transmit signal 225. Each receiver thus must first demultiplex
and filter the received signal to determine or extract for further
processing only the carrier of the source signal or uplink channel
that is directed to the particular terminal. In a further
embodiment, in the case of a multi-channel receiver, as would be
recognized, the receiver may determine and process multiple carrier
frequency signals of multiple uplink channels directed to the
particular terminal. For simplicity, however, the following
description addresses example embodiments encompassing a
single-channel receiver. Accordingly, for example, the receiver
203a will first process the received transmission signal 225 to
isolate the carrier phase and frequency of the uplink signal 223a.
In that regard, the receiver 203a includes the receiver section
231a. In one embodiment, the receiver section may comprise a bank
of receiver mixers to frequency/phase-translate each carrier of the
received signal 225, where the translation may be expressed as
exp(-j(2.pi.f.sub.1t+.theta..sub.1))/ {square root over (M.sub.c)}
for the signal 223a, and generally as
exp(-j(2.pi.f.sub.mt+.theta..sub.m))/ {square root over (M.sub.c)}
for the m.sup.th signal 223. The signal then passes through the
receive filter bank 232a, and through the processing of the
receiver section 231a and the filter bank 232a, the receiver 203a
extracts the carrier signal 223a, effectively tuning to the carrier
frequency and phase of the uplink signal directed to the particular
receiver. In one embodiment, the input-output relationship of the
receive filter bank 232 of the m.sup.th receiver 201 may be
expressed as:
x.sub.m(t)=.intg..sub.-.infin..sup..infin.r(t) {square root over
(M.sub.c)}exp(-j(2.pi.f.sub.mt+.theta..sub.m))P.sub.m,R(t-.lamda.)d.lamda-
. (5)
where m=1, . . . M.sub.c, and .lamda. is an integration
variable.
[0046] The filter bank 232a (e.g., P.sub.m,R(t)) models the cascade
of the matched filter and the on-board output multiplexing (OMUX)
filter of the satellite transponder. The outputs of the receive
filter bank are then sampled at the FTN symbol rate of the data
source 205a to allow for fractionally-spaced equalization. An
IMUX/OMUX equalizer 233a (e.g., a fractionally-spaced equalizer for
IMUX/OMUX equalization) is then employed to compensate for the
linear distortion resulting from the IMUX/OMUX filters of the
satellite transponder 217. In other words, the IMUX/OMUX filter
compensates for the linear distortion or ISI produced by the IMUX
and OMUX filters of the satellite transponder 217, while passing
the non-linear interference of the HPA and the linear interference
of the FTN rates for compensation by the Turbo Volterra Module
235a.
[0047] According to one embodiment, based on the known
characteristics of the IMUX and OMUX filters of the satellite
transponder, the IMUX/OMUX equalizer 233 can be configured to train
for the appropriate compensation. For example, during a
non-operational training period, a sequence of known symbols can be
transmitted over the respective satellite channel, and the
IMUX/OMUX equalizer can train for the appropriate compensation for
the linear interference reflected by the received signals. Further,
as the characteristics of the IMUX and OMUX filters aboard the
satellite may change with temperature or over time, the IMUX/OMUX
equalizer can periodically retrain to update the compensation based
on changing operating conditions or the passage of a particular
length of time (e.g., taking into account the degree of change over
time of such onboard filters).
[0048] According to a further embodiment, based on the known
characteristics, the IMUX/OMUX equalizer can apply compensation
based on an inversion of the known group delay of the IMUX and OMUX
filters. Then, at the output of the IMUX/OMUX equalizer 233a, the
signal y.sub.m(n.tau.T.sub.S) is generated by sampling the output
of the IMUX/OMUX equalizer at the transmission symbol rate of
1/.tau.. For example, because the extracted version of the received
carrier signal reflects a sequence of source symbols, each mapped
to a corresponding or respective constellation point, the signal is
sampled in synchronization with the sample rate of the source
symbols as transmitted (e.g., at the FTN rate) to obtain a received
form or representation of each of the respective transmitted signal
constellation points.
[0049] According to example embodiments (as depicted in FIG. 2),
each receiver 203 employs a Turbo Volterra Module 235 to mitigate
the impact of the FTN-induced linear ISI as well as the HPA-induced
nonlinear distortion, where soft-information is exchanged between
the FEC decoder and a Volterra filter within the Turbo Volterra
Module. FIG. 7 illustrates a block diagram of the Turbo Volterra
Module of the receivers 203 of FIG. 2, in accordance with example
embodiments of the present invention. The Turbo Volterra Module 235
comprises a log-likelihood ratio (LLR) computation module 711, a
de-interleaver module 713, a soft-in/soft-out (SISO) decoder module
715, a hard decision module 717, an interleaver module 719, a
bit-to-symbol mapper module 721, a Volterra filter module 723, and
a subtractor module 725.
[0050] The Turbo Volterra Module 235 is configured to reconstruct
both the distortion caused by the HPA of the downlink transponder
217 (which is essentially nonlinear) and the linear ISI due to FTN
signaling rate. According to example embodiments, the LLR
computation module 711 converts the input signal 731 into an
updated set of soft information 735, to match the distortion
exhibited by the received signals and thereby facilitate improved
performance of the decoder module 715. By way of example, based on
LLRs 740 (L.sub.a.sup.(E)(c.sub.m,n')) from the previous decoding
iteration, the LLR computation module converts the input signal 731
into extrinsic LLRs 735 (L.sub.e.sup.(E)(c.sub.m,n')) regarding the
interleaved code bits. The updated soft information 735 is then
de-interleaved by the de-interleaver module 713 to form the LLRs
737 (L.sub.a.sup.(D))(c.sub.m,n')), and decoded by the decoder
module 715 to generate the LLRs 739 (L.sup.(D)(c.sub.m,n')). The
LLRs 739 are then interleaved by the interleaver module 719 to form
the LLRs 740 (L.sub.a.sup.(E))(c.sub.m,n')). The signal conversion
by the LLR computation module 711 is performed through a
predetermined number of inner loop iterations (the inner LLR
computation loop of signal
735.fwdarw.737.fwdarw.739.fwdarw.740.fwdarw.735) for each iteration
of the outer decoding loop.
[0051] With regard to the outer decoding loop, the interleaved LLRs
740 (L.sub.a.sup.(E))(c.sub.m,n')) are also provided for the next
decoding iteration (the outer decoding loop of signal
739.fwdarw.740.fwdarw.741.fwdarw.743.fwdarw.731.fwdarw.735.fwdarw.737.fwd-
arw.739). In that regard, because the actual transmitted symbols
are effectively unknown, the decoder module 715 decodes the LLRs
737 to determine soft estimates of the transmitted symbols, which
are provided by the decoder in the form of soft information LLRs
739 (L.sup.(D)(c.sub.m,n')). The soft information LLRs 739 are
interleaved to generate the LLRs 740
(L.sub.a.sup.(E))(c.sub.m,n')), and are then passed through an
estimation or model of the channel (applied through the
bit-to-symbol mapper module 721 and the Volterra filter module
723). In that regard, the soft information LLRs 740 are
bit-to-symbol mapped via the bit-to-symbol mapper module 721, and
are then provided to the Volterra filter module 723 as a set of
LLRs 741 (L.sub.a.sup.(E)(a.sub.m,n)). The Volterra filter module
723 then estimates the interference based on the bit-to-symbol
mapped LLRs 741. In other words, based on the soft information LLRs
739, the Volterra filter module thereby estimates the nonlinear
interference exhibited by the received signal. The interference
estimate of the Volterra filter module is then subtracted from the
received signal by the subtractor 725 to generate an improved
version of the received signal (an improved estimate of the
transmitted or source signal based on the received signal minus the
estimated nonlinear interference) for a next iteration. In other
words, compensation for the distortion is accomplished by
subtracting the interference estimate 743 (provided by the Volterra
filter module 723) from the input signal 729, to obtain the
estimated signal 731.
[0052] Accordingly, for each outer loop decoding iteration, the
soft information LLRs are improved (based on the interference
estimate from the prior iteration), which leads to an improved
signal estimate. In turn, the improved signal estimate leads to
further improved soft information LLRs for the next outer loop
decoding iteration. In other words, with each decoding iteration,
the estimate of the nonlinear interference is improved, which in
turn improves the estimate of the transmitted signal (e.g., based
on the received signal minus the improved estimate of the nonlinear
interference). Further, the decoder module 715 may comprise any
form of SISO decoder (e.g., a low density parity check code (LDPC)
decoder) configured to generate the soft information LLRs for
feedback to a Volterra filter in such an iterative manner to
provide improved estimates for compensation of the interference
effects manifested by the received signal. Moreover, because the
Volterra representation is sparse and the significant terms are
few, the representation can be truncated (insignificant terms can
be dropped) without a significant degradation in performance, which
facilitates a more efficient and less complex receiver
implementation.
[0053] More specifically, during the first iteration, in the
absence of the soft information LLRs 739 provided by the FEC
decoder from prior iterations, the estimate signal 731 ({tilde over
(y)}(n.tau.T.sub.s)) input to the LLR computation module 711 is
considered as being equal to the sampled signal 729
(y(n.tau.T.sub.s)) input to the Turbo Volterra Module 235. The LLR
computation module 711 converts the input signal 731 into the soft
information 735. The resulting soft information 735 is then
de-interleaved and provided as the LLRs 737, to be used by the SISO
decoder module 715. The SISO decoder module 715 decodes the updated
soft information 737 to determine the estimates of the transmitted
symbols (in the form of the soft information LLRs 739), which are
provided as feedback to compensate for the nonlinear interference
(e.g., the nonlinear distortion resulting from the HPA of the
satellite transponder) and the linear interference (e.g., the ISI
resulting from the FTN transmission symbol rate) in the input
signal, thereby providing an improved signal 731 ({tilde over
(y)}(n.tau.T.sub.s)) for the next iteration. The soft information
LLRs 739 are interleaved and bit-to-symbol mapped, and are then
provided to the Volterra filter module as the set of LLRs 741. The
Volterra filter module 723 computes the expectation
E{a.sub.NL.sup.(3)(n)|L.sub.a.sup.(E)}, as follows:
E { a _ NL ( 3 ) ( n ) | L a ( E ) } = i = n = - ( L - 1 ) / 2 n +
( L - 1 ) / 2 E { a m , i v m , i ( a m , i v m , i * ) * | L a ( E
) } ( 6 ) ##EQU00004##
The parameters v.sub.m,i and v*.sub.m,i arise from the Volterra
filter characterization of the resulting distortion, as shown in
equation (16), below. As an example, from equation (16), it is
apparent that when i=n-(L-1)/2, v.sub.m,i=2 and v*.sub.m,i=1. The
individual terms in the product shown above are then computed
as:
E { a m , i v m , i ( a m , i v m , i * ) * | L a ( E ) } = l = 1 M
a l v m , i ( a l v m , i * ) * P { a l = a l | L a ( E ) } ( 7 )
##EQU00005##
where the conditional symbol probability is formed using the soft
information LLRs 741 (L.sub.a.sup.(E)(a.sub.m,n)). Compensation for
the distortion is accomplished by subtracting the interference
estimate 743 (provided by the Volterra filter module 723) from the
input signal 729 (y(n.tau.T.sub.s)), to obtain the estimate signal
731 ({tilde over (y)}(n.tau.T.sub.s)), such that:
{tilde over
(y)}(n.tau.T.sub.s)=y(n.tau.T.sub.s)-[h(n)E{a.sub.NL.sup.(3)(n)|L.sub.a.s-
up.(E)}-E{.rho..sup.centroid(a.sub.m,n|L.sub.a.sup.(E))}] (8)
where h(n) is the set of coefficients of the Volterra filter
(chosen to model the HPA and FTN induced distortion), and
.rho..sup.centroid(a.sub.m,n) is the centroid associated with the
symbol a.sub.m,n of the samples at the input of the LLR computation
module 711.
[0054] Then, for the next iteration, the LLR computation module 711
utilizes the improved signal 731 ({tilde over (y)}(n.tau.T.sub.s))
for conversion into the soft information 735. As with the prior
iteration, the resulting soft information 735 is de-interleaved and
provided as the LLRs 737, to be used by the SISO decoder module
715. The SISO decoder module 715 decodes the updated soft
information 737 to determine improved estimates of the transmitted
symbols (in the form of the soft information LLRs 739), which are
provided as feedback to for improved compensation for the nonlinear
interference in the input signal (based on the results of the prior
iteration), thereby providing a further improved signal 731 ({tilde
over (y)}(n.tau.T.sub.s)) for the next iteration. The soft
information LLRs 739 are interleaved and bit-to-symbol mapped, and
are then provided to the Volterra filter module as a new set of
LLRs 741. The Volterra filter module 723 computes a new expectation
or estimate, and improved distortion compensation is accomplished
by subtracting the improved interference estimate 743 (resulting
from this new iteration) from the input signal 729.
[0055] Accordingly, the operation of the Turbo Volterra Module 235
is based on a number of inner iterations of the LLR computation
module 711 (e.g., inner loop iterations of the LLR computation loop
735.fwdarw.737.fwdarw.739.fwdarw.740.fwdarw.735) for each iteration
of the outer decoding loop
(739.fwdarw.740.fwdarw.741.fwdarw.743.fwdarw.731.fwdarw.735.fwdarw.737.fw-
darw.739). The number of inner iterations for the LLR computation
module and the number of outer iterations for the feedback loop are
predetermined numbers based on a tradeoff between system complexity
and the desired level of performance. The numbers for the inner and
outer iterations are predetermined values, which can be obtained
through system simulations. As the iterations are increased, the
increase in performance reflected by the simulation results will be
outweighed by the added complexity of the interference compensation
process--the increase in complexity reaches a point of diminishing
returns.
[0056] According to an example embodiment, the Volterra filter
module 723 is configured to operate in the following manner. For
the m.sup.th carrier, h.sub.m (n) can be defined as:
h _ m ( n ) = .DELTA. [ h _ m ( 1 ) ( n ; L ' ) h _ m ( 3 ) ( n ; L
) ] where , ( 9 ) h _ m ( 1 ) ( n ; L ' ) = .DELTA. [ .eta. _ ( 1 )
( n ; L ' ) ] and ( 10 ) h _ m ( 3 ) ( n ; L ' ) = .DELTA. [ .eta.
_ ( 3 ) ( n ; L ) ] ( 11 ) ##EQU00006##
The vector h.sub.m.sup.(1) (n; L') is in turn composed of vector
.eta..sup.(1)(n; L'), which incorporates the memory of the
1st-order interference of size L' symbols, such that:
.eta. _ ( 1 ) ( n ; L ' ) = .gamma. ( 1 ) [ h m ( 1 ) ( ( L ' - 1 2
) T s ) h m ( 1 ) ( ( L ' - 1 2 - 1 ) T s ) h m ( 1 ) ( ( - L ' - 1
2 ) T s ) ] ( 12 ) ##EQU00007##
and the first-order Volterra kernel is:
h m ( 1 ) ( t ) = .intg. - .infin. .infin. p m , T ( t - .tau. ) p
m , R ( .tau. ) .tau. ( 13 ) ##EQU00008##
The vector h.sub.m.sup.(3)(n; L) is in turn composed of vector
.eta..sup.(3)(n; L), which incorporates the memory of the 3rd-order
interference of size L symbols, such that:
.eta. _ ( 3 ) ( n ; L ) = .gamma. ( 3 ) [ h m ( 3 ) ( ( L - 1 2 ) T
s , ( L - 1 2 ) T s , ( L - 1 2 ) T s ) h m ( 3 ) ( ( L - 1 2 ) T s
, ( L - 1 2 ) T s , ( L - 1 2 - 1 ) T s ) h m ( 3 ) ( ( - L - 1 2 )
T s , ( - L - 1 2 ) T s , ( - L - 1 2 ) T s ) ] ( 14 )
##EQU00009##
and the third-order Volterra kernel is:
h m ( 3 ) ( t 1 , t 2 , t 3 ) = .intg. - .infin. .infin. p m , T (
t - .tau. ) p m , T ( t 2 - .tau. ) p m , T * ( t 3 - .tau. ) p m ,
R ( .tau. ) .tau. ( 15 ) ##EQU00010##
Accordingly, the Volterra filter comprises two components, a first
order component and a third order component. The first order
component of the Volterra filter handles the linear interference
(e.g., the linear ISI attributable to the FTN symbol transmission
rate), and the third order component handles the nonlinear
interference (e.g., the nonlinear distortion attributable to the
satellite transponder HPAs).
[0057] According to one example embodiment h.sub.m(n) can be
computed analytically using .gamma., h.sub.m.sup.(1)(t), and
h.sub.m.sup.(3)(t.sub.1, t.sub.2, t.sub.3). According to a further
example embodiment, h.sub.m(n) is instead determined using
stochastic gradient-based algorithms to iteratively derive the
solution without a priori knowledge of the kernels. By way of
example, the corresponding vector of 1st and 3rd-order symbol
combinations in a.sub.NL.sup.(3) (of equation (8), above) are
expressed as:
[ a m , n - L ' - 1 2 a m , n + L ' - 1 2 a m , n - L - 1 2 a m , n
- L - 1 2 a m , n - L - 1 2 * a m , n - L - 1 2 a m , n - L - 1 2 a
m , n - L - 1 2 + 1 * a m , n + L - 1 2 a m , n + L - 1 2 a m , n +
L - 1 2 * ] ( 16 ) ##EQU00011##
[0058] The proposed receiver of example embodiments thereby
maintains a complexity that is not exponential with the alphabet
size M, which is particularly useful with FTN-induced distortion,
which tends to linger over large number of symbols.
[0059] According to example embodiments, with regard to the
operation of the LLR computation module 711, the module 711 may be
configured to determine a plurality of likelihood metrics (LMs),
which in turn are used to generate log-likelihood ratios (LLRs) to
be passed to the decoder 237 for the determination of the code-bits
of the respective transmitted source data symbol. According to one
embodiment, for example, each LM may be based on a sample
representation with respect to a one signal constellation point
(with respect to the corresponding source data symbol), and a
different one of a plurality of core parameters (CPs), where each
CP is based on a centroid estimate with respect to a different
signal cluster. According to a further embodiment, each LM may be
based on the sample representation with respect to the one signal
constellation point (with respect to the one corresponding source
data symbol), and a different one of a plurality of variance
parameters (VPs), where each VP is based on a variance estimate
with respect to a different signal cluster. According to yet a
further embodiment, each LM may be based on the sample
representation with respect to the one signal constellation point
(with respect to the one corresponding source data symbol), and a
different one of a plurality of correlation parameters (CnPs),
where each CnP is based on a correlation estimate with respect to a
different signal cluster. Further, the LLR computation module 711
may be configured to determine the LMs based on the sample
representation with respect to the one signal constellation point,
along with a combination of one or more of the CPs, CnPs and VPs.
Such operational methods for the LLR computation module 711 are
further described in copending U.S. patent application Ser. No.
13/622,348 (filed 18 Sep. 2012), which is incorporated herein in
its entirety. Alternatively, the LLR computation module 711 may
employ other techniques for the determination of likelihood metrics
(LMs) and/or log-likelihood ratios (LLRs) for facilitating improved
operation of the decoder module.
[0060] The following provides a performance evaluation with respect
to various example embodiments, based on an extensive Monte-Carlo
simulation study. The simulation results reflect performance
results based on: (1) transmitter and receiver implementations as
illustrated in FIGS. 2 and 7; (2) carriers that are non-overlapping
in frequency (e.g., frequency spacing is (1+.alpha.)R.sub.s; (3)
transmit and receive filters P.sub.m,T(t) and P.sub.m,R (t) being a
matched pair of root-raised cosine (RRC) filters with a roll-off
factor of 0.10; and (4) the forward error correction (FEC) being
LDPC encoding and decoding with an LDPC code of codeblock length
64800 bits. Further, the performance charts reflect a DVB-S2
standard system as a benchmark for illustrating the improvement in
terms of spectral efficiency measured in bits/sec/Hz. The spectral
efficiency may be defined as (bits/sec/Hz):
.eta. = R c log 2 M ( 1 + .alpha. ) .tau. ##EQU00012##
[0061] FIG. 8 illustrates packet error rate (PER) performance
curves (as a function of the per-symbol signal-to-noise ratio
(SNR)) for an example system employing IMUX and OMUX filters and a
TWTA high power amplifier within the transponders of the satellite,
wherein, in each case, the system achieves a spectral efficiency of
2.42 bps/Hz, and each curve reflects application of a particular
modulation and coding scheme in combination with either a prior art
least mean square (LMS) adaptive equalizer or a Turbo Volterra
Module (of example embodiments of the present invention) employed
within the receiver. The curve 811 illustrates the performance
achieved by 8PSK modulation with the DVB-S2 LDPC code at code rate
8/9, employing an LMS equalizer in the receiver, and without
employing FTN rates. The maximum number of LDPC decoder iterations
is set to 50. Alternatively, the curve 813 illustrates the
performance achieved by a 1+7APSK modulation scheme, with the
DVB-S2 LDPC code at code rate 8/9, employing an LMS equalizer in
the receiver, and without employing FTN rates. As shown by this
curve, the capacity limitations arising from using the DVB-S2 8PSK
constellation (which puts 8 uniformly spaced constellation points
on a single ring) can be overcome by using a 1+7APSK constellation
and bit-to-symbol labeling and constellation symbol positioning as
specified in the following table (Table 1). For the same alphabet
size (M=8), as shown by the curve, the 1+7APSK constellation
provides a significant performance improvement over the DVB-S2 8PSK
modulation example.
TABLE-US-00001 TABLE 1 Bit Label [x, y] Coordinates 000 [0.0, 0.0]
001 [{square root over ((8.0 * .epsilon..sub.x/7.0)}, 0.0] 010
[{square root over (8.0 * .epsilon..sub.x/7.0)} * cos(4.0 *
.pi./7.0), {square root over (8.0 * .epsilon..sub.x/7.0)} * sin(4.0
* .pi./7.0)] 011 [{square root over (8.0 * .epsilon..sub.x/7.0)} *
cos(2.0 * .pi./7.0), {square root over (8.0 * .epsilon..sub.x/7.0)}
* sin(2.0 * .pi./7.0)] 100 [{square root over (8.0 *
.epsilon..sub.x/7.0)} * cos(12.0 * .pi./7.0), {square root over
(8.0 * .epsilon..sub.x/7.0)} * sin(12.0 * .pi./7.0)] 101 [{square
root over (8.0 * .epsilon..sub.x/7.0)} * cos(10.0 * .pi./7.0),
{square root over (8.0 * .epsilon..sub.x/7.0)} * sin(10.0 *
.pi./7.0)] 110 [{square root over (8.0 * .epsilon..sub.x/7.0)} *
cos(6.0 * .pi./7.0), {square root over (8.0 * .epsilon..sub.x/7.0)}
* sin(6.0 * .pi./7.0)] 111 [{square root over (8.0 *
.epsilon..sub.x/7.0)} * cos(8.0 * .pi./7.0), {square root over (8.0
* .epsilon..sub.x/7.0)} * sin(8.0 * .pi./7.0)]
[0062] The curve 815 illustrates the performance achieved by the
1+7APSK modulation scheme, with the DVB-S2 LDPC code at code rate
8/9, and employing a Turbo Volterra Module (in accordance with
embodiments of the present invention) in the receiver, but without
employing FTN rates. Further, the curve 817 illustrates the
performance achieved by now applying a 16APSK modulation scheme,
with the DVB-S2 LDPC code at code rate 2/3, employing an LMS
equalizer in the receiver, and without employing FTN rates. The
16APSK modulation scheme uses 4-bits per symbol, but achieves
better performance using a stronger low rate code (2/3)--the extra
energy level of the 16APSK constellation, however, adds a
significant level of complication with regard to the nonlinear
distortion. Lastly, the curve 819 illustrates the performance
achieved by the foregoing 1+7APSK modulation scheme, with the
DVB-S2 LDPC code at code rate 5/6, and employing both a Turbo
Volterra Module (in accordance with embodiments of the present
invention) in the receiver and an FTN rate of 6.67%. As illustrated
by the curve 819, in accordance with embodiments of the present
invention (employing an FTN symbol transmission rate and a Turbo
Volterra Module in the receiver), a 1+7APSK modulation scheme can
be maintained, with a lower rate code (5/6 which is lower than the
8/9 code rate), with improved performance--an approximately 1.5 dB
improvement over the 8PSK and 8/9 code rate combination of the
DVB-S2 standard, and an approximately 0.4 dB improvement over a
16APSK and 2/3 code rate combination of the DVB-S2 standard.
Accordingly, the advantages provided by example embodiments of the
present invention facilitate the employment of an B-ary modulation
constellation with a stronger low rate FEC code (using an FTN
symbol transmission rate), while maintaining a desired level of
spectral efficiency and improving system performance at the same
time.
[0063] Moreover, while system performance is generally affected by
the particular bit labeling and bit positioning for each
constellation, the optimal labeling and bit positions specified in
Table 1 are not unique in that certain specific modifications of
bit labeling and bit positioning can achieve equivalent
performance. One such modification exists with respect to the bit
positions, whereby equivalent performance can be achieved with a
1+7APSK signal constellation as specified by Table 1, but where
each of the [x, y] bit positions is rotated by a fixed rotation
factor (e.g., each bit position is rotated by the same rotation
factor, such as 5 degrees, 7 degrees, 12 degrees, etc.). Other
modifications exist with respect to the bit labeling, whereby
equivalent performance can be achieved with a 1+7APSK signal
constellation as specified by Table 1, but where the bit labeling
is modified by interchanging the 0's and 1's (changing each one to
a zero and changing each zero to a one in each bit label) and/or by
applying a uniform swapping of bit positions within each bit label
(uniformly swapping one or more bit positions with one or more
corresponding other bit positions in each bit label--e.g., swapping
the first and third bit label positions within each bit label).
Moreover, any of the foregoing specific modifications can either be
applied by itself or in combination with any one or more of the
other specific modifications.
[0064] FIG. 9 illustrates a computer system upon which example
embodiments according to the present invention can be implemented.
The computer system 900 includes a bus 901 or other communication
mechanism for communicating information, and a processor 903
coupled to the bus 901 for processing information. The computer
system 900 also includes main memory 905, such as a random access
memory (RAM) or other dynamic storage device, coupled to the bus
901 for storing information and instructions to be executed by the
processor 903. Main memory 905 can also be used for storing
temporary variables or other intermediate information during
execution of instructions to be executed by the processor 903. The
computer system 900 further includes a read only memory (ROM) 907
or other static storage device coupled to the bus 901 for storing
static information and instructions for the processor 903. A
storage device 909, such as a magnetic disk or optical disk, is
additionally coupled to the bus 901 for storing information and
instructions.
[0065] According to one embodiment of the invention,
implementations of an interference compensation system and
algorithms, in accordance with example embodiments, are provided by
the computer system 900 in response to the processor 903 executing
an arrangement of instructions contained in main memory 905. Such
instructions can be read into main memory 905 from another
computer-readable medium, such as the storage device 909. Execution
of the arrangement of instructions contained in main memory 905
causes the processor 903 to perform the process steps described
herein. One or more processors in a multi-processing arrangement
may also be employed to execute the instructions contained in main
memory 905. In alternative embodiments, hard-wired circuitry is
used in place of or in combination with software instructions to
implement the embodiment of the present invention. Thus,
embodiments of the present invention are not limited to any
specific combination of hardware circuitry and software.
[0066] The computer system 900 also includes a communication
interface 917 coupled to bus 901. The communication interface 917
provides a two-way data communication coupling to a network link
919 connected to a local network 921. For example, the
communication interface 917 may be a digital subscriber line (DSL)
card or modem, an integrated services digital network (ISDN) card,
a cable modem, or a telephone modem to provide a data communication
connection to a corresponding type of telephone line. As another
example, communication interface 917 may be a local area network
(LAN) card (e.g., for Ethernet.TM. or an Asynchronous Transfer Mode
(ATM) network) to provide a data communication connection to a
compatible LAN. Wireless links can also be implemented. In any such
implementation, communication interface 917 sends and receives
electrical, electromagnetic, or optical signals that carry digital
data streams representing various types of information. Further,
the communication interface 917, for example, includes peripheral
interface devices, such as a Universal Serial Bus (USB) interface,
a PCMCIA (Personal Computer Memory Card International Association)
interface, etc.
[0067] The network link 919 typically provides data communication
through one or more networks to other data devices. For example,
the network link 919 provides a connection through local network
921 to a host computer 923, which has connectivity to a network 925
(e.g., a wide area network (WAN) or the global packet data
communication network now commonly referred to as the "Internet")
or to data equipment operated by service provider. The local
network 921 and network 925 both use electrical, electromagnetic,
or optical signals to convey information and instructions. The
signals through the various networks and the signals on network
link 919 and through communication interface 917, which communicate
digital data with computer system 900, are example forms of carrier
waves bearing the information and instructions.
[0068] The computer system 900 sends messages and receives data,
including program code, through the network(s), network link 919,
and communication interface 917. In the Internet example, a server
(not shown) might transmit requested code belonging to an
application program for implementing an embodiment of the present
invention through the network 925, local network 921 and
communication interface 917. The processor 903 executes the
transmitted code while being received and/or store the code in
storage device 239, or other non-volatile storage for later
execution. In this manner, computer system 900 obtains application
code in the form of a carrier wave.
[0069] The term "computer-readable medium" as used herein refers to
any medium that participates in providing instructions to the
processor 903 for execution. Such a medium may take many forms,
including but not limited to non-volatile media, volatile media,
and transmission media. Non-volatile media include, for example,
optical or magnetic disks, such as storage device 909. Volatile
media may include dynamic memory, such as main memory 905.
Transmission media may include coaxial cables, copper wire and
fiber optics, including the wires that comprise bus 901.
Transmission media can also take the form of acoustic, optical, or
electromagnetic waves, such as those generated during radio
frequency (RF) and infrared (IR) data communications. Common forms
of computer-readable media include, for example, a floppy disk, a
flexible disk, hard disk, magnetic tape, any other magnetic medium,
a CD ROM, CDRW, DVD, any other optical medium, punch cards, paper
tape, optical mark sheets, any other physical medium with patterns
of holes or other optically recognizable indicia, a RAM, a PROM,
and EPROM, a FLASH EPROM, any other memory chip or cartridge, a
carrier wave, or any other medium from which a computer can
read.
[0070] Various forms of computer-readable media may be involved in
providing instructions to a processor for execution. For example,
the instructions for carrying out at least part of the present
invention may initially be borne on a magnetic disk of a remote
computer. In such a scenario, the remote computer loads the
instructions into main memory and sends the instructions over a
telephone line using a modem. A modem of a local computer system
receives the data on the telephone line and uses an infrared
transmitter to convert the data to an infrared signal and transmit
the infrared signal to a portable computing device, such as a
personal digital assistance (PDA) and a laptop. An infrared
detector on the portable computing device receives the information
and instructions borne by the infrared signal and places the data
on a bus. The bus conveys the data to main memory, from which a
processor retrieves and executes the instructions. The instructions
received by main memory may optionally be stored on storage device
either before or after execution by processor.
[0071] FIG. 10 illustrates a chip set 1000 in which embodiments of
the invention may be implemented. Chip set 1000 includes, for
instance, processor and memory components described with respect to
FIG. 9 incorporated in one or more physical packages. By way of
example, a physical package includes an arrangement of one or more
materials, components, and/or wires on a structural assembly (e.g.,
a baseboard) to provide one or more characteristics such as
physical strength, conservation of size, and/or limitation of
electrical interaction.
[0072] In one embodiment, the chip set 1000 includes a
communication mechanism such as a bus 1001 for passing information
among the components of the chip set 1000. A processor 1003 has
connectivity to the bus 1001 to execute instructions and process
information stored in, for example, a memory 1005. The processor
1003 includes one or more processing cores with each core
configured to perform independently. A multi-core processor enables
multiprocessing within a single physical package. Examples of a
multi-core processor include two, four, eight, or greater numbers
of processing cores. Alternatively or in addition, the processor
1003 includes one or more microprocessors configured in tandem via
the bus 1001 to enable independent execution of instructions,
pipelining, and multithreading. The processor 1003 may also be
accompanied with one or more specialized components to perform
certain processing functions and tasks such as one or more digital
signal processors (DSP) 1007, and/or one or more
application-specific integrated circuits (ASIC) 1009. A DSP 1007
typically is configured to process real-world signals (e.g., sound)
in real time independently of the processor 1003. Similarly, an
ASIC 1009 can be configured to performed specialized functions not
easily performed by a general purposed processor. Other specialized
components to aid in performing the inventive functions described
herein include one or more field programmable gate arrays (FPGA)
(not shown), one or more controllers (not shown), or one or more
other special-purpose computer chips.
[0073] The processor 1003 and accompanying components have
connectivity to the memory 1005 via the bus 1001. The memory 1005
includes both dynamic memory (e.g., RAM) and static memory (e.g.,
ROM) for storing executable instructions that, when executed by the
processor 1003 and/or the DSP 1007 and/or the ASIC 1009, perform
the process of example embodiments as described herein. The memory
1005 also stores the data associated with or generated by the
execution of the process.
[0074] While example embodiments of the present invention may
provide for various implementations (e.g., including hardware,
firmware and/or software components), and, unless stated otherwise,
all functions are performed by a CPU or a processor executing
computer executable program code stored in a non-transitory memory
or computer-readable storage medium, the various components can be
implemented in different configurations of hardware, firmware,
software, and/or a combination thereof. Except as otherwise
disclosed herein, the various components shown in outline or in
block form in the figures are individually well known and their
internal construction and operation are not critical either to the
making or using of this invention or to a description of the best
mode thereof.
[0075] In the preceding specification, various embodiments have
been described with reference to the accompanying drawings. It
will, however, be evident that various modifications may be made
thereto, and additional embodiments may be implemented, without
departing from the broader scope of the invention as set forth in
the claims that follow. The specification and drawings are
accordingly to be regarded in an illustrative rather than
restrictive sense.
* * * * *